43

Chapter -II

Approaches towards Code Mixing and Code Switching

Sociolinguistics considers Code Mixing and Code Switching as the by-

products of bilingualism. In Code-Mixing, a fluent bilingual changes the

language by using words from other language without any change at all in

situation, whereas in Code Switching, anyone who speaks more than one

language chooses between them according to circumstances and according to

the language comprehensive to the persons addressed , the purpose is to get the

right effect of communication. With this primary understanding, we will

consider these phenomena in some detail.

2.1 Nature and Scope

2.1.1 Two Different Views

At the outset, it is necessary to mention that, as per the available literature

on CM and CS there are two different views about maintaining the distinction

between CM and CS. Some scholars like Kachru (1983), Annamali (1989),

Bokamba (1988), Sridhar and Sridhar (1980), Hamers and Blanc (1990), Bhatia

(1992), Poplack (1980) treat these phenomena as the distinct manifestations.

Some other scholars like Eastman (1992), Scotton (1992), however, consider

that there is no distinction between them.

2.1.1.1 In Favour of Maintaining Distinction

Kachru (1983:193) observes, “There is a distinction between CM and CS,

though they have been treated as the language contact phenomenon. The CS

entails the ability to switch from code A to code B. The alteration of codes is

determined by the function, the situation and the participants. It refers to

categorization of one‟s verbal repertoire in terms of functions and roles. The

CM, on the other hand, entails transferring linguistic units from one code into

another.”

Bokamba (1989), while maintaining difference in CM and CS, notes three

points:

44

1) The two phenomena must be distinguished, because each makes a different

linguistic and psycholinguistic claim. For instance, CS does not necessitate

the interaction of the grammatical rules of the language pair involved in the

speech event, whereas CM does.

2) CM exemplifies the most advanced degree of bilingualism to the extent that it

requires considerable competence in the simultaneous processing of the

grammatical rules of the language pair [cf. Kachru (1978, 1982 a), Sridhar

and Sridhar (1980), Poplack (1990), Sankoff and Poplack (1981) and

Bokamba (1988)]. Linguistically and perceptually, therefore, CS and CM

cannot be constructed as the co-variant phenomena. The degree of

bilingualism implied in the production of the code-mixed sentences suggests

that only highly proficient bi/multilingual can successfully engage in

sustained code-mixed production.

3) CM typically involves the use of two languages at a time, although

occasionally three are used. Regardless of the number of languages involved

in the discourse, the language that provides the grammatical structure into

which elements are inserted is referred to as the host while the other is termed

the guest language. (Sridhar and Sridhar, 1980).

2.1.1.2 Against Maintaining Distinction

According to Scotton (1992), the borrowed and code-switched forms

behave in the same way morphosyntactically in the matrix language, hence

should not be seen as distinct processes. Eastman (1992:1) notes that the urban

language contact phenomena do not distinguish CM, CS and Borrowing. The

urban settings where people from diverse linguistic and cultural backgrounds

regularly interact make it abundantly clear that in normal everyday conversation,

material from many languages may be embedded in a matrix language regularly

and unremarkably.

Though there are different views about the distinction between Code

Mixing and Code Switching, some linguists, have used CS as the cover term to

refer to these two phenomena. There are some others like Muysken (1997), who

use these terms interchangeably. Muysken (2000) uses CS for alteration; Bhatia

45

(1992) on the contrary, uses CM as cover term for CM and CS. Clyne (2003:75)

says, “We should reserve Code Switching for transference of individual lexical

items through to whole stretches of speech; but we should adopt different terms

like transversion for cases where the speaker crosses over completely into the

other language.”

Though, Code Switching, as Gardner-Chloros (2009) points out, has

gained wider currency in the language interaction phenomenon, on the whole,

among most of the linguists, there has been a general agreement on maintaining

the distinction. The discussion that follows further, elaborate this.

2.1.2 Definitions of Code Mixing

Many linguists and scholars have tried to define CM in their own way.

According to Mary W.J. Tay (1989:408), “Code-mixing involves the embedding

or mixing of various linguistics units, i.e. morphemes, words, phrases and

clauses from two distinct grammatical systems or sub-systems within the same

sentence and same speech situation.”

Kachru (1978:28) uses this concept to refer to “the use of one or more

languages for consistent transfer of linguistic units from one language into

another and by such a language mixture developing a new restricted or not

restricted code of linguistic interaction.” Bokamba (1989:278) notes, “Code

mixing is the embedding of various linguistic units such as affixes (bound

morphemes), words (unbound morphemes) phrases and clauses from two distinct

grammatical (sub) systems within the same sentence and speech event. That is,

CM is an intrasentential switching.” Weinreich (1963:1) uses the term

„Interference Phenomena‟ to imply the rearrangement of patterns that result from

the introduction of foreign elements into the more highly structured domains of

language, such as the bulk of the phonemic system a large part of the

morphology and syntax and some areas of vocabulary.

Crystal (1997:66), in „The Dictionary of Linguistics and Phonetics‟,

defines code-mixing as a linguistic behaviour that “involves the transfer of

linguistic elements from one language into another.” Gumperz Hermandez-

46

Chavez (1978) notes, “Code-mixing is a type of borrowing, where depending on

various linguistic factors speakers borrow items of various sizes.”

For Hudson (1996:53), CM takes place „where a fluent bilingual talking

to another fluent bilingual changes the language without any change at all in the

situation.‟ The purpose of CM seems to be to symbolize a somewhat ambiguous

situation for which neither language on its own would be quite right. To get the

right effect, the speakers balance the two languages against each other as a kind

of linguistic cocktail, a few words of one language, then a few words of the

other, then back to the first for a few more words and so on. The changes

generally take place more or less randomly as for as subject matter is concerned,

but they seem to be limited at structural level.

2.1.3 Definitions of Code Switching

Hudson (1996:53) discusses CS as the „inevitable consequences of

bilingualism, as any one who speaks more than one language chooses between

them according to circumstances.‟ Annamalai (1989:48) observed, “Switching is

normally done for the duration of a unit discourse. A bilingual speaker can

switch between mixed codes as he does between unmixed languages. Switching

is found only in a balanced and stable bilingual.”

Crystal (1995) states, “Code or language switching occurs when an

individual, who is bilingual, alternates between two languages during his or her

speech with another bilingual person.” Halliday (1978:65) defines CS as „code-

shift actualized as a process within the individual: the speaker moves from one

code to another and back, more or less rapidly, in course of a single sentence.‟

Verma (1976:156) focuses on Code Switching as „a verbal strategy used

by speakers in much the same way as creative artists switch styles and levels (i.e.

from sublime to the mundane or the serious to the comic or the vice versa) or the

ways in which monolinguals make selections from among vocabulary items‟ and

concludes, “Each type of coding or CS is appropriate to the topical and

situational features that give rise to it.” Weinriech (1953:73) elaborates, “The

ideal bilingual switches from one language to another according to appropriate

47

changes in speech situation, but not in unchanged speech situation and certainly

not within a single sentence.”

Bokamba (1989:278) considers CS as „the mixing of words, phrases and

sentences from two distinct grammatical (sub) systems across sentence

boundaries within the same speech event, in other words, intersentenstial

switching.‟ Ashok Kumar (1995:44) adds, “Code Switching which is influenced

by extra-linguistic factors such as topic, interlocutors, setting etc. is the alternate

use of lexical items, phrases, clauses and sentences from the non-native language

into the system of the native language.” Gardner Chloros (2009) is of the opinion

that CS may be used as „a general term covering all outcomes of contact between

two varieties whether or not there is evidence of conversions.‟ For her, Code

Switching refers to the use of several languages or dialects in the same

conversation or sentence by the bilingual people.

Regarding the precise definitions of CM and CS, it appears that it is

extremely difficult to maintain the difference between CM and CS for all

purposes. As CM/CS is a growth area in linguistics and as it is being studied

from several different perspectives, it may be said that, one definition covers

only one aspect/perspective of the huge phenomena of CM/CS. In this regard

one may agree with what Milroy and Muysken (1995 in Gardner Cholros

2009:12) noted in the introduction to One Speaker Two languages about the

confusing structure of CM/CS. They assert, “The field of CS research is replete

with a confusing range of terms descriptive of various aspects of phenomenon.

Sometimes referential scope of these terms overlaps and sometimes particular

terms are used in different ways by different writers.” One observation in this

regard is also right which reads, “CM/CS term distinction should be treated as

research tool while describing the data only.”

2.1.4 Code Mixing, Code Switching and Borrowing

CM and CS are further compared by linguists with Borrowing.

Borrowing is maintained as related to CM and CS, but also different from

them.

48

2.1.4.1 Borrowing as Different from CM/CS

Kamwangamalu (1992) explains this phenomenon in detail. He notes,

“Structurally, unlike CM and CS, Borrowing entails integration of linguistic

units from one language into the linguistic system of other language. The

linguistic units thus integrated become part of the linguistic system of the

borrowing language they take on its phonological, morphological and syntactic

characteristics, and enter into its lexicon [e.g. Gumperz (1982)]. Functionally,

unlike CM and CS, Borrowing sometimes occurs to fill lexical gaps in the

lexicon of the borrowing language.” Moreover, Borrowing may occur in the

speech of monolingual and bilingual speakers alike, whereas CM and CS occur

in the speech of bilingual speakers only. He presents these features in a table

format (1992:174). Regarding this format, he notes that it is adapted from

current works on the phenomena under consideration and in particular from the

works of Kachru (1978), Sridhar and Sridhar (1980), Gumperz (1982) and

Poplack (1990). The characteristics of CM, CS and Borrowing listed in the

following table, are by no means exhaustive. They are simply indented to

provide a background against which we can better categorize the data to be

analyzed.

Table -3: Salient Features of CM, CS and Borrowing

Salient features CM CS B

Linguistic units involved are essentially sentences - + -

Require speakers to share the same code repertoire + - -

Sometimes occur to fill lexical gaps in L1 - - +

Require/imply changes (e.g. of topic, setting, participants) in the speech

situation - + -

Occur to mark confidentially or an aside for explanation - + -

Sometimes occur to mark eliteness and /or modernness + + +

Presuppose bilingual competence of the speaker but not necessarily of the

hearer - + -

Entails integration (phonological, syntactic, and lexical) of L2 material in L1 - - +

Treated as part of L1. - - +

Lexicon L1 equivalent exits. + + -

Wide diffusion in the community n.a n.a +

Accepted as one‟s own L1 items n.a n.a +

49

Hudson (1996) differentiates Borrowing from CM/CS on the basis of

speech and language systems. He notes, “Whereas Code Switching and Code

Mixing involved mixing language in speech, Borrowing involves mixing the

systems themselves because an item is „borrowed‟ from one language to become

part of other language. For example - words for foods, plants, instructions, music

and so on which most people can recognize as borrowings (or Loan –Words).”

SalMons (1990) uses four central criteria to distinguish Borrowing from

Code Switching, following Scotton (1988):

A. Frequency of occurrence is often the most central criterion. The frequent

items tend to be considered Borrowings, the infrequent ones more likely

represent code switches or nonce borrowings.

B. Phonological integration is taken as the indication of Borrowing and lack

of integration is understood as an indication of CS or nonce borrowings.

C. Syntactic integration, as with phonological integration, is taken as the

indication of Borrowing; with lack of syntactic integration, it is

understood as an indication of CS.

D. Lexically, Pfaff (1982:269-273) gives several directly relevant criteria for

drawing a distinction when dealing with items of unclear status.

The relation between Code Mixing, Code Switching and Borrowing is

extensively reviewed by Romaine (1989) and Myers-Scotton (1992) also.

Boschoten (1998) points out that though any aspect of a language including its

structures can be borrowed, the importance of this relation lies in the fact that

„single word code-switches/loans are the commonest kind of Code Switching in

many situations.‟ As Poplack, Sankoff and Miller (1988; 62) note, „within single

words, common nouns are the most frequency borrowed items.‟ This observation

is supported by three kinds of explanation. One is by Bynon (1977:231) who

views it as reflecting the size of grammatical categories concerned. Another is

suggested by Aitchison (2000:62), as the nouns are freer of syntactic restrictions

than other word-classes. The third, which is a more sociolinguistic one, is that

these are accessible to the bilinguals with any degree of competence, even

50

minimal, in the language from which the borrowing is taken (Gardner-Chloros,

2009).

2.1.4.2 Myers Scotton’s Views on Borrowing and Code Switching

Myers Scotton (1992) claims that CS and Borrowing are „universally

related process.‟ According to her Matrix Language Frame (MLF) Model, CS

occurs everywhere within a frame which is set by the matrix language. The term

matrix refers to the language in which the majority of morphemes in a given

conversation occur. The languages from which material enters a matrix language

are referred to as embedded. Central to MLF theory is the idea that content

(often nouns, verbs, etc.) and system (articles, inflections etc) morphemes in the

embedded language are accessed differently by the matrix language. Myers

Scotton is concerned with finding out which singly occurring embedded

language lexemes are borrowings and which are code-switches, in light of the

fact that both borrowed and code-switched forms behave the same way

morphosyntactically in the matrix language. Her MLF model posits structural

constraints which account for both borrowings and code switches.

She uses data from Kenya and Zimbabwe to explain her model. Swahili is

the matrix language of Kenya and Shona is the matrix language of Zimbabwe.

Her Matrix Language Frame Model provides a principled basis for considering

both Borrowing and Code Switching processes to be a part of a single

continuum. She explores four hypotheses:

1. Matrix language + embedded language [EL] constituents conform to

the morpheme order of the matrix language and in ML+EL

constituents articles and inflections (System morphemes with non-

lexical information) come from matrix language.

2. The EL content morphemes that are not congruent with stored lemmas

(lexical information in a mental lexicon are blocked from appearing in

ML+EL constituents. (This is called Blocking Hypothesis)

3. Embedded language forms which cannot be accounted for by the

blocking hypothesis are subject to an embedded language trigger

hypothesis.

51

4. The peripheral and formulaic embedded language constituents may

occur with relative freedom in a matrix language. ( This is called EL

Hierarchy Hypothesis)

On the continuum, the forms range from abrupt loans (e.g. the term

school fees as used in shone language) to core borrowings (e.g. the term weekend

in Swahili language), to actual instances of CS. According to Myers-Scotton,

frequency is the best criterion to use to link borrowings to the mental lexicon in

contrast to single form code switches. Yet, there is little reason to make a

distinction between the processes. Neither morphosyntactic nor phonological

integration criteria remain viable ways to decide whether embedded language

material is the result of borrowing or code switching. For example the word

shule (school) is a long established borrowing into Swahili (originally from

German); however, the verb stem-visit and the noun difference are considered

code switching forms, largely because they show no frequency of occurrence

and occur only in one conversation in a set of 40 of the data.

The MLF model of CS which can account for single form code switches

also accounts for borrowed lexical items. Single or multi-word code switches are

retained in the mental lexicon of the embedded language, are lemma entries

there, while borrowings continue to remain matrix language lexical items. All

code switches structurally represent material embedded into a matrix language

while all borrowings are matrix language material par excellence.

In accordance with this model, the English preposition at, for example, is

blocked from appearing in the Swahili sentence nilikuwa nataka kumpata

stadium which translates as „I wanted to find him stadium‟ rather than „I wanted

to find him at the stadium. At (or for that matter the as well) has no counterpart

in Swahili, so stadium „occurs as a bare form‟ making it look as much like a

borrowing as a code-switch. In contrast, other English prepositions do have

congruent forms in Swahili and may occur themselves as single code-switched

forms (kati ya and between so match).

52

2.1.4.3 Borrowing and CS as Continuum

Shaffer (1978), while dealing with the place of CS in linguistic contacts,

discusses „interlingual impact‟ defined and studied by Weinreich (1963), Clyne

(1967) and Haugen (1956). He accepts Haugen‟s three stages of diffusion,

namely CS, interference, and integration. As per this scheme, the switching

constitutes the alternate use of the distinct codes; interference constitutes

overlapping and integration constitutes interlingual impact only in historical

sense.

According to Haugen‟s scheme, interference and integration have in

common a certain degree of leveling in the structural distinctiveness of both

codes. By contrast, CS involves the stringing together of „unadopted‟ words and

phrases. [Haugen (1956:40), Haugen (1973:528)]. Interference is contrary to

contemporary norms, whereas integration is in harmony with current norms.

The acceptance of Haugen‟s diffusion concept is further modified by

different suggestions: Interference will be confined to imperfect second

language acquisition. So, interference will be recognized as restricting the

native-like distinctiveness of the two codes available to bilingual. Thus, there

appears to be a continuum between switching and borrowing. It also seems that

some times switches are gradually borrowed. There would then be roads to

lexical borrowing, instaneous borrowing (as in the case of sputnik, borrowed

from Russian overnight) and gradual integration of the lexical items used in

frequent switches.

2.1.4.4 Borrowing and CM/CS in Urban Setting

According to Eastman (1992:1), the multilingual phenomenon in urban

setting does not distinguish CM, CS and Borrowing. The urban settings where

people from diverse linguistic and cultural backgrounds regularly interact makes

it abundantly clear that in normal everyday conversation, material from many

languages may be embedded in a matrix language regularly and unremarkably.

At the same time, forms may be embedded noticeably. Where people use a

mixed language regularly, code switching represents the norm (an unmarked

choice after Myers Scotton 1983). Where people invoke another language is an

53

obvious way, position of relative social, political or economic strength is often

being negotiated and Code Switching represents a marked choice. In addition, it

is seen that the absence of CS in the multilingual urban contents may be

indicative of the tacit acceptance of political, social, and/or economic power

differences.

2.2 Syntactic Constraints

The syntactic constraints in CM/CS have been studied and commented

upon time to time by the linguists while dealing with CM/CS phenomena.

The formal and theoretical oriented studies on Code Mixing started

growing in the decade of 1970 with Tim (1975), McClure (1977), Lipski (1978),

Pfaff (1979), Poplack (1980) and others. In the decade of 1980, the systematic

attempts have been made by Wool Ford (1983) and Di Sciullo et al (1986) to

develop a formal model of code-mixing within transformational-generative and

government-binding frameworks, respectively. These models attempt to explain

why some syntactic elements are subject to code-mixing while others are

unlikely or impossible candidates for mixing. The models then go on to show

that either the shared portion of the participating grammars (phrase structure

rules etc.) or principles such as government set the stage for language mixing.

Bhatia‟s (1989) study on bilinguals‟ creativity and syntactic theory,

however, draws two conclusions. One, the conception of the grammar of CM

envisioned in the formal studies is too narrow in nature; and two, there is a

pressing need for reexamination of such notions as „hybrid rules‟ and „new

grammar ‟. Evidence has been presented which supports the conclusions that a

new grammar of Filmi-English is motivated by structural properties of Hindi and

English principles of Universal grammar, whereas the rule of „o’ insertion in the

derivation of infinitival verbs is motivated by the consideration of informal

discourse. An attempt has been made to present identifiable and distinguishing

properties of a hybrid grammatical system and a new grammatical system.

Finally, it was shown that code-mixing does not represent a surface level

phenomenon.

54

2.2.1 Possibility of Universal Constraints

Annamali (1989), in his research paper, „Language Factor in Code

Mixing‟, studies the universal constraints on CM and proposes that universal

constraints are likely to exist if CM is a universal phenomenon. He means to say

that any bilingual can mix his languages and any two languages can be mixed.

He assumes that the mixed code has the properties of a natural language. The

assumed linguistic properties of the mixed code, according to him, are- (a)

Mixing is governed by levels of units such as word level, clause level, and so on;

(b) It is sensitive to syntactic constituents like noun phrase, verb phrase, verb

phrase, etc. and (c) It is a variable with reference to word classes such as nouns ,

verbs, etc. In other words, it has organizational property, configuration property

and classificatory property. There are constraints which prohibit mixing within a

word or a phrase which illustrates. There are constraints which prohibit partial

mixing across constituents which illustrate. There are constraints which stipulate

that mixing will follow a hierarchy of word classes, which illustrates, a) for

example, if verbs are mixed, nouns would be mixed and mixing will include

more items in the noun class than in the verb class. Further, the distinction

between CM and CS is made in this study with regard to such constraints.

Mixing, for example, is the linguistic strategy for discourse functions primarily

involving social meanings and switching is a discourse strategy for linguistic

(verbal) communication reflecting language competence and preferred language

choice of the participants.

The switching is found only in a balanced and stable bilingual, while

mixing appears in incipient and attrited bilinguals as well. The mixing is largely

motivated by the need to fill gaps in the linguistic competence of the speaker.

The constraints on mixing are not likely to be the same in stable, incipient and

attrited bilinguals.

Further, by adopting the terminology such as MT for Mother Tongue and

OT for Other Tongue, the universal constraint of mixing in two types of

bilinguals is proposed. There are two types of bilinguals- balanced and

imbalanced. According to him, the balanced bilingual mixes his MT with OT

55

more than the imbalanced bilingual. The imbalanced bilingual mixes with low

frequency only with nouns.

Hudson (1996) is of the view that the syntactic categories used in

classifying linguistic items may be independent of their social descriptions. At

least some syntactic (and other) categories used in analyzing language are

universal rather than tied to particular languages. He firmly states that there is

no doubt that there are syntactic constraints; people who belong to code-mixing

communities can judge whether particular constructed code-mixed examples are

permitted or not, and these judgments are on the whole borne out by studies of

texts.

2.2.2 Bokamba’s Negative Views on Syntactic Constraints

Bokamba (1989) also treats this „Syntactic Constraint‟ aspect of CM.

According to him, “conceptually, the basic assumption underlying almost all

syntactic studies of CM is that the occurrence of code-mixed speech among bi

and multilingual is fundamentally differ from the occurrence of other speech

variants or registers. Hence, ungrammaticalities in the code-mixed speech are

best treated as violations of syntactic constraints of the language-pair involved in

the discourse rather than as violation of the morpho-syntactic rules governing

that particular dialect. The code-mixed speech is characteristically and

definitionally a dialect of the host language concerned and postulation of

Syntactic Constraints to account for such speech varieties is both inadequate and

misconceived.

This study claims that there are no general or universal surface syntactic

constraints on CM, because these changes are naturally called for by the very

nature of language as a linguistically structured and socio-psychologically rule –

governed means of communication. The failure to discover and establish such

constraints, it has been argued, is due to the narrow focus attributed to syntactic

research on CM where crucial socio-psychological factors are left out of

consideration. The postulation of the surface syntactic constraints, regardless of

the scope of their validity (e.g. language specific of putative universal), is not

only misguided but also unwanted at this stage of the research.

56

2.2.3 CS and Syntactic Constraints at Lexical level

Hussein (1993) studies the CS of Arabic-English bilinguals. It describes

various syntactic constraints that govern this phenomenon in their speech, the

susceptibility of particular sentence constituents to switching and the

determination of the matrix language in their linguistic performance.

CS among the Arabic-English bilinguals, like other language contact

situations, supports the fact that this phenomenon is not random but rule

governed. The verbal behavior of the respondents utilized for this study clearly

suggests certain linguistic constraints that not only determine the switches that

occur in their speech but also the acceptability of these switches by the

bilinguals.

While it is believed that extra-linguistic factors are the only means that

motivate bilinguals to switch code styles, or registers, this study has attempted to

deliberate syntactic restrictions in the speech of the Arabic English bilinguals.

Pronominal subjects, objects and their verbs are considered the main restrictions

against switching between Arabic and English; the pronominal subjects and

objects and their verbs are tightly linked in a way that they have to adhere to the

linguistic rules of a certain code rather than the mixture of two codes.

On the basis of the manner of CS experienced by the respondents, it is

observed that the Arabic, the mother tongue of these bilinguals, is invariably the

matrix language. In these sentences demonstrating CS, the matrix clause is in

Arabic. Moreover, the English constituents in Arabic clauses are deemed to

adapt to the morphological and phonological rules governing Arabic. Finally,

this study points to the fact that both languages are independent and are unlikely

to move towards merging; the linguistic constraints are, in fact, intrinsic to both

languages and this an overwhelming evidence that each language is keeping its

linguistic integrity at least for the foreseeable future.

Timm (1975) deals with the several constraints on Code Switching. The

first category deals with switching of pronominal subjects or objects that were

considered possible, although not common. It was noted, however, that

expressions such as they were chopeando, he was cahado, etc. and English

57

loanwords which had been morphologically adapted to fit Spanish paradigms

were possible.

Finally, within noun phrases, there is considerably more flexibility as

regards to the possibilities of CS, although certain restrictions do appear. The

data, which have been surveyed, suggest that the true model of bilingual

competence may lie somewhere between the two diametrically opposite poles,

being neither two completely distinct language systems nor one homogeneous

amalgam. Those bilingual speakers capable of engaging in spontaneous CS

apparently posses the ability to mentally compare equivalent sentences in the

two languages for degree of syntactic compatibility and to code-switch only in

those instances where such compatibility would be preserved.

2.2.4 Syntactic Constraints and Psychological Reasons

Sridhar and Sridhar (1980) investigate the syntax of Code Mixing in

psycholinguistic perspective. The syntactic constraints dealt by them have the

consequences for a psychological model of bilingual information processing.

They present the „Dual structure Principle‟ which is refinement of Poplack‟s

(1980) Equivalence Constraints and Hypothesis of Borrowing of Lipski

(1978).This principle is stated as follows:

Dual Structure Principle: The internal structure of the guest constituents

need not conform to the constituent‟s structure rules of the host language, so

long as its placement in the host sentence obeys the rules of the host language

(Sridhar and Sridhar 1980: 209).The authors arrive at the following conclusions

regarding its psycholinguistic aspects:

a) The entire rule system of both the languages, not just their respective

lexicons are simultaneously active when mixed sentence types are

produced. It rules out the possibility of a „single switch model‟ during

the code-mixing process.

b) The activation of one language system does not result in the non-

operation of the other.

58

c) The two languages systems are neither „merged‟ together, nor are they

totally independent of each other. Rather, they interact in complex

ways to produce complex code-mixed patterns.

2.2.5 Poplack’s Views on Syntactic Constraints

The paper by Sankoff and Poplack (1981) proposes a formal grammar of

CS, accepting at the same time that the complete understanding of CS could only

be achieved through combined ethnographic, attitudinal and grammatical study,

i.e. an integrated analysis not only of when people code-switch but also how

where and why (Sankoff and Poplack 1981:4).

Two constraints of language alteration between Spanish and English

(„free morphemes constraint‟ and „equivalence constraint‟) are posited. These

constraints serve as the basis for the formal description of the syntax of Spanish-

English Code Switching. The universality of these constraints is doubted, since it

is held, and rightly so, that they may not operate with reference to English and

some highly inflected or agglutinative language (e.g. Japanese and Turkish) or

language with an actively different word-order (e.g. Hindi), because the

constraints on CS are based on bilingual performance not transformationally

generated. The authors believe in direct generation of CS sentences by a context-

free grammar. The CS constraints are surface phenomena and can not naturally

be generated in deep structure. Phrase Structure Grammar of L1 and L2 can be

combined to from a Code Switching Grammar which generates grammatical

monolingual sentences as well as those containing only valid code-switches.

(Sankoff and Poplack 1981: 36).

2.2.6 Pfaff’s Views on Syntactic Rules in CM

Pfaff (1979) has considers the several aspects of Spanish/English

intrasentential mixing. She found that speakers who code-switch are competent

in the syntactic rules of both languages. She claims that the third grammar does

not exist to account for the utterances in which the languages are mixed. Instead,

the grammar of Spanish and English are meshed according to the given

constraints: functional and lexical constraints, structural constraints, semantic

constraints and discourse constraints.

59

2.3 Discourse Markers in CM/CS

Discourse Markers are the responses that indicate text as a conversation.

Generally, they are in one word or two words. They do not have independent

identity, but are very meaningful when they occur in conversation. O.K., Yes,

No, I mean, of course, yeah, ouch, woo, nei, weel, you know are such English

responses. In general, the educated Indian Speaker, while conversing in English,

uses these English responses. However, it is observed that most of the times, he

switches to his mother tongue only for the similar responses. For example,

„Come on yaar, do not take it seriously‟ or „we are going for shopping, you are

coming na?‟ In these examples, the speaker switches only for yaar and na. The

CS is essentially a product of bilingual or multilingual phenomena and one of

the major concerns of Sociolinguistics is to study the behavior of languages in

such phenomena.

2.3.1 Functions of Discourse Markers in Code Mixing and Code Switching

Discourse Markers are defined by Schifrin (1987) as „contextual

coordinates of talk‟. They display coherence; some markers have referential or

expressive meaning, others have interactional meaning and all markers may have

various functions in the text. Though Schifrin (1987) suggests that many

Discourse Markers are multifunctional, one can classify discourse markers

according to their primary function; of course, their specific function depends on

the context in which they are used. In her study, the Discourse Markers are

multifunctional. They fall into the following categories:

1) Participation category (Y’ know and I mean): They have little semantic

content serve as an interactional function.

2) Connectives (and and but) and markers of cause or result (because and

so), both of which have semantic content.

3) and and so functioning not as connectives or markers of a certain

relation between clauses, but as fillers and clause initiators and thus

with more interactive than semantic value.

Sal Mons (1990:453) reviews the role of Discourse Markers in code-

switching that changes into Borrowings. Discourse Markers such as „well’ and

60

„you know’ in English have been considered as „extra-sentenced or emblematic‟

code-switching (Poplack, 1982). In American German varieties spoken in central

Texas (and other areas), English markers are used in conversation to the normal

exclusion of the German discourse-marking system.

English Discourse Markers themselves do not represent proper code-

switching in his study, but instead have the status of borrowings (in line with

recent discussions of that distinction), while the discourse making systems of

German and English have undergone convergence, This conclusion is relevant to

other bilingual situations and could entail reclassifying such phenomena as

borrowings or convergence, where these have been assumed to represent

switches. His analysis also underscores the necessity of including „convergence‟

in addition to borrowings and code-switching in the study of

bilingual/multilingual communities.

Gumperz (1982) treats such kind of switching as an integration or

sentence filler. This occurs when languages merge into one another in language

contact situation.

The studies mentioned above help us to arrive at the conclusion that the

switched Discourse Markers are extra-sentenced, emblematic, sentence fillers,

contextual coordinates having participatory value. Most of the markers are

borrowed and become one with the common code. They are multifunctional and

their specific function depends on the context.

2.4 Functional Aspects

Several studies have dealt with the functions of CM and CS. These

functions vary as per the status of language pair involved, socio-cultural

divergences and the situation in which they occur. It is also noted that CS

functions as a symptom of quite opposite developments. It occurs in language

accommodations, in language divergence, in language maintenance and in

language shift. It reflects social differences and tendencies within the same

society and language combination. This aspect is dealt in the studies of Li

Wei,(1998a). On the contrary, in the studies of Poplack,(1988) and

61

McClure(1998), CS is dealt with social differences and tendencies between

different societies and different language combinations.

2.4.1 Gumperz’s Observations on Functions of CS

Gumperz (1982:75-76) mentions various types of functions of CS:

1) Quotations: In many instances, the code-switched passages are clearly

identifiable either as direct quotations or as reported speech. Where the

quotations were originally in a code-mixed variety, they also appear in reported

speech in the same variety.

2) Addressee specification: This function explains switches which serve to direct

the message to one of several possible addressees.

3) Integrations: According to this function, the code switch serves to mark an

integration or sentence filler. In the data from Singapore, numerous particles are

found in conversation irrespective of what is the predominant language used.

These are the results of various languages merging in contact with one another

and evolving a common code with a common vocabulary. To treat such

particles as mere interjections or sentence fillers seems counterintuitive.

4) Reiteration: Frequently, a message in one code is repeated in other code either

literally or in somewhat modified form. In some cases, such repetitions may

serve to clarify what is said but often they simply simplify or emphasis a

message.

5) Message qualification: A large group of switches consists of qualifying

constructions such as sentence and verb complements or predicates following a

copula. In such cases, the main message is in language X, whereas a

qualification of the message is given in language Y.

6) Personalization vs. objectivization: This dichotomy suggests a difference

between talk about action and talk as action, the degree of speaker‟s involvement

or distance from a message, whether a statement reflects personal opinion or

knowledge, or whether it refers to specific instances or has the authority of a

generally known fact.

62

2.4.2 CM as Mark of Modernization

Kamwangamalu (1989) demonstrates that CM is a cross-cultural

phenomenon. To support his claim, he considers the functional use of CM,

attitudes towards CM and the language change as the result of CM.

From a functional point of view, he focuses on CM as a mark of

modernization. The use of CM for modernization is attested to in a variety of

domains, for example, education, literature, science and technology. He points

out that the use of CM is a character feature of the speech of the elite group or

those who hold a higher socioeconomic status in their respective communities.

CM is both to these speakers and the aspiring masses a model of language use

with which to identify.

In attitudinal terms, it has been shown that, although CM proves to be the

norm of speech most observed in bilingual communities around the world, the

bilinguals do not always admit it. This is because CM is considered by some as a

„corrupt‟, an „impure‟ linguistic behavior. If such contact takes place, there will

be language change as a result of CM.

He further refers to Hymes‟ observation that “the fact that English, a

former immigrant language… has not prevented other immigrant language (e.g.

Italian, Spanish, French and German) from not only being spoken, but also

mixed with English. In that respect, CM sheds light on the current life of

language in bilingual communities around the world.”

2.4.3 Code Mixing as Communicative Strategy

Tay (1989) focuses on the use of CM and CS as a communicative strategy

in multilingual communities among proficient bilingual speakers. She uses the

spontaneous conversations which involve code-switching and mixing between

some of the major languages in Singapore, such as English, Mandarin, Hokkien.

She stresses that, the typical code switcher or mixer is usually not aware of why

he/she switches codes at certain points of the discourse; and, to try to evolve a

functional typology to fit all situations, would, therefore, seem futile. Instead, it

is suggested that total communicative impact created by the discourse should be

looked as discourse and ask questions such as: Is the speaker trying to establish

63

rapport or is he distancing himself? She emphasizes that the communicative

intent of the bilingual speakers is of „prime importance‟; once we get the

intention then, we might then be able to look at the various strategies used by

them and how codes are manipulated to achieve these purposes. According to

her, the effectiveness of a communicative strategy does not depend on the choice

of the code as such but considerations such as- which code has the most colorful,

expressive, shortest, most economic way of repeating or elaborating upon what

was said earlier.

With regard to the study of the new varieties of English, she notes that,

these varieties should not be treated as deviant forms of some native variety, but

as independent systems. She has observed in this study that the treatment of

Singaporean English as a learner variety is unsatisfactory, because it fails to

capture the essential difference in motivation between two very different types

of CM, one motivated by inadequate vocabulary on the part of genuine learner

and the other motivated by the desire to communicate as effectively and

expressively as possible on the part of the fluent bilingual. About Singaporean

English, her observation is that it is far richer than, for example, British English

because of its contact with other languages spoken in Singapore. Instead of

merely documenting purely formal or purely functional characteristics of

bilingual speech, there is the urgent need to look at bilingual speech as primarily

an act of communication. Thus, those functions which are „communicative‟

rather than act of communication (ritualistic) will be most relevant to the study

of bilingualism. Similarly, form is to be perceived as the outcome of

communication rather than the focus of bilingual speech.

2.4.4 Functions of Code Mixing in New Varieties of English

D‟Souza (1992) deals with the relationship between CM and the New

Varieties of English (NVEs). According to him, the NVEs are varieties of

English that have emerged as a result of the colonial experience. This fact is

important, because it contributes a great deal to making the NVE‟s what they

are. The NVEs have the following characteristics: They are the result of

colonization, are instutionalized, have range and depth, are nativized and stable,

64

have developed through the educational systems in bilingual context and are

creative.

Given these characteristics, one has to realize that in talking of an NVE one

is not talking of a monolith but of a code that encompasses several sub-varieties.

The NVEs then, like the Old Varieties of English, have a range of styles and

registers and regional variation. But in addition, they have code-mixed styles and

registers which are also so intrinsic to them that CM has to be added to the list of

characteristics that define the NVEs. CM may be used for the following reasons

in the NVEs:

a) To conceal speakers‟ regional caste/religious identity.

For example, Annamali (1978) notes that Tamil CM with English may

be used to conceal caste identity when the term „brother-in-law‟ is used

instead of „maccan’ or „attimbeer’.

b) For register identification

c) To impress one‟s interlocutors and force inferior status on others while

claiming prestige for oneself. In the NVE countries, English is the

language of Prestige and CM with English is used to impress.

d) To prove education, modernity urbanity etc. (Sridhar 1978:10)

e) As a neutralizing device: For example, in Kashmir, the word mond

carries the annotations that the English equivalent „widow‟ does not

(Kachru 1986).

D‟Souza questions the widespread assumption that CM is an essential

feature of the NVEs. It claims that CM is the result of bilingualism and

language contact and will be found in all such contexts even when English has

become a world language and one result of its spread is that it plays an important

part in mixing. The mixed varieties arise even in non-NVE contexts. To tie CM

to the study of the NVEs is to blind ourselves to many crucial aspects of both

phenomena and to claim that NVEs are in some essential way different from

other codes.

65

2.4.5 CS and Preference of the Speaker

Chengappa (2005) has attempted to relate the amount of CS to the

proficiency and preference of the speaker. The significance of the paper is in

that, it is prepared by using the Matrix Language Frame given by Myers Scotton

(1993 b). It makes distinction between the Western bilingualism and the Indian

bilingualism as it notes, “The bilingualism in the western world is not at a grass

root level as in India and the other Asian and African countries. So it is difficult

to generalize results obtaining from the western studies into the Indian Context.

The variety and the use of English spoken in India are also different and this

may in turn affect the CS evidenced in the Indian context. It is noted that the

operating mode of the speaker affects the amount and type of CS in the

experimental situation. The operating mode of the bilingual changes from one

instance to the other, depending on the community in which he is in.” Hence,

the study tries to correlate the speech mode of a bilingual with the amount of

CS seen. It also aims at comparing the CS across monolingual and the bilingual

communicative situations.

2.5 Socio-cultural and Pragmatic Factor

Most of the scholars are of the opinion that the sociolinguistic factor is

the key to understand why Code-switching takes the form it does in each

individual case. This sub-section summarizes the studies of the scholars

concentrating on the sociolinguistic and the pragmatic aspects of CM and CS.

2.5.1 Gardner-Chloros on CS and Sociolinguistic Factors

Gardner-Chloros (2009:42) discusses three types of factors contributing

to the form taken by CS in a particular instance:

1) The factors those are independent of particular speakers and

circumstances in which the varieties are used, which affect all the

speakers of the relevant varieties in particular community- e.g.

economic „market‟ forces- covert prestige, power relations, and the

associations of each variety, with a particular context or way of life.

66

2) The factors that are attached to the speakers, both as individual and as

members of a variety of subgroups- their competence, their social

networks and relationships, their attitudes and ideologies, their self

perception and perception of other.

3) The factors that are within the conversations where CS takes place.

CS provides bilinguals with tools to structure their discourse beyond

those available to the monolinguals.

The other three sociolinguistic determinants of CS pointed out by her are:

1) Conversational/Pragmatic Motivations:

Myres-Scotton (1993a:49) deals with the distinctions between the

allocational paradigm in which the social structure determines the

language behavior, and the interactional one in which individuals

make rational choices to achieve their goals.

Milroy and Gordon (2003) similarly make distinction between CS

which exploits the symbolism or connotations of each of the ends and

CS that purely exploits the context which the two varieties provide.

Gumperz (1982) makes the distinction between we code and they

code, While dealing with the pragmatic function of CS.

2) Socio-psychological influences:

Accommodation, attitude and audience design are the concepts used by

the socio-psycholinguists. CS is one of the possible ways of

accommodating to the interlocutors linguistic preferences. It can serve

as a compromise between two varieties, where these carry different

connotations or social meanings for speakers and in the locators.

This compromise function is not limited to the spontaneous

speech. It is also exploited by politicians in their speeches as comedians

in their jokes. Generally, it is used in the media which aim at

multilingual audiences for multiple functions.

3) Addresses Specification

Another major function of CS is what Gumperz called „addresses

specification‟. The use the appropriate language to address different

67

interlocutors allows the participants to continue the conversation

smoothly.

„Gender‟ is also considered as one of the most important

sociolinguistic categories. CS connects with gender issues.

2.5.2 CS Between Different Social Settings

Bloom and Gumperz (1972:409) have discussed the use of Ranamal (a

rural district variety in Northern Norway) and Bokmal (the standard variety) as

the symbols of local and national Norwegian social values. This study shows

that it is possible to find patterns in the phonological, grammatical and semantic

choices made by an individual, and predict their occurrence with the help of

certain social factors. Therefore, the claim that there is a constant relationship

between the linguistic and the social structures is not without basis. The authors

examine the linguistic repertoire of the Hemnesberget community and report the

formal and the functional distribution of the Ranamal and Bokmal. The analysis

proves that distinction with regard to the socio-economic status of

Hemnesbergests, their descent (local vs. non-local) and their use of and loyalty

towards the dialect or the standard, tend to operate simultaneously. CS, in this

setting, is considered to be narrowly constrained by the observable features of

the situation (locale, participants, etc.) and also as a linguistic force capable of

defining and altering that situation. The authors here make a basic distinction

between a „situational‟ type of CS which is governed by situational norms, and

„metaphorical‟ CS where alteration enriches a situation , allowing for allusions

to more than one social relationship within the situation.

Gumperz (1964:1117) studies Hindi–Punjabi CS and observes, “The need

for a number of individuals tends to reduce the language distance between codes.

Linguistic overlap is the greatest in those situations which favor inter-group

contact.” One reason for this distinctiveness is that, “they have coexisted with

the same linguistic area as part of the same cultural complex for several hundred

years”. The result, which Gumperz arrives at, indicates the differences in the

speech of Punjabi-Hindi bilinguals to be mostly grammatical.

68

2.5.3 Conversational Code Switching

Gumperz (1982) focuses on the central issues of the CS phenomenon. His

(1982) paper deals with „conversational Code Switching‟ involving three

language pairs: Spanish and English (Sp-E), Hindi and English (H-E): and

Slovenian and German (Sl-G) in South Western United States, Northern India

(Delhi) and Austria respectively. Gumperz here aims at investigating the

communicative aspects of CS and “to show how speakers and listeners utilize

sub-consciously internalized social and grammatical knowledge in interpreting

bilingual conversation” (63-64). In other words, the study implies a need for

exploring the strategies that determine the meaning potential of CS and the

connection that holds between them and the socio-linguistic knowledge that

speakers and listeners share among themselves. The important observations of

this paper are:

1) CS, as a type of speech variation, reflects the role of verbal signals in

human interactions;

2) CS signals contextual information equivalent to what in monolingual

settings is covered through prosody or other syntactic or lexical processes.

It generates the presuppositions in terms of which the content of what said

as decoded. The speakers‟ notion of a code is personalized and hence

differs from that of the linguist. The speakers depend on these notions to

relate their grammatical system to the extra-linguistic environment.

(Gumperz 1982:98)

2.5.4 CS and Extra-linguistic Factors

Weinreich (1963: 3) while discussing „Interference phenomenon‟ says,

“The precise effect of bilingualism on a person‟s speech varies with a great

many other factors, some of which might be called extra-linguistic because they

lie beyond the structural differences of the languages or even their lexical

inadequacies.” Among the non-structural factors, he notes that some are inherent

in the bilingual person‟s relation to the languages he brings into contact. These

factors are:

69

a) the speakers facilitating of verbal expression in general and his ability to

keep two languages apart,

b) the relative proficiency in each language,

c) the specialization in the use of each language by topics and interlocutors,

d) the manner of learning each language,

e) the attitudes toward each language- whether idiosyncratic or stereotyped,

f) the size of bilingual group and its socio-cultural homogeneity or

differentiation- its breakdown into sub-groups using one or the other

language as their mother tongue, demographic facts, social and political

relations between these sub-groups,

g) the prevalence of bilingual individuals with given characteristics of

speech behavior in several sub-groups,

h) the stereotyped attitudes towards each language „prestige‟, the indigenous

or immigrant status of the languages concerned,

i) the attitudes toward the culture of each language community,

j) the attitudes towards bilingualism as such,

k) the tolerance or intolerance with regard to mixing languages and to

incorrect speech in each language,

l) the relation between the bilingual group and each of two language

communities of which it is a marginal segment,

Weinreich believes that it is in broad psychological and socio-cultural

settings that the effects of bilingualism can best be understood.

2.5.5 Language Choice as per Situation

Hudson (1996) maintains the view that the choice of language is

controlled by the rules. The members of the community learn these rules from

their experience; so, these rules are the part of their total linguistic knowledge.

For example, according to Denison (1971), everyone in the village of Sauris, in

northern Italy, spoke German within the family, Saurian (a dialect of Italian)

informally within the village, and standard Italian to outsiders and in more

formal village settings (school, church, work). Because of this linguistic

division of labour, each individual could expect to switch codes (i.e.

70

languages) several times in the course of a day. This kind of code-switching is

called situational code-switching because the switches between languages

always coincide with changes from one external situation (for example, talking

to members of the family) to another (for example, talking to the neighbours).

The rules link the languages to different communities (homes, Sauris,

Italy), so each language also symbolizes that community. Speaking the

standard Italian at home would be like wearing a suit and the speaking German

in the village would be like wearing beach-clothes in church. In short, each

language has social functions which no other language could fulfill. These

social functions are less arbitrary results of history; but they are no less real for

that. The same seems to be typical of the bilingual communities in general.

The main reason for preserving the language is because of the social

distinctions that they symbolize. In clear cases, we can tell what situation we

are in just by looking around us; for example, if we are in lecture–room full of

people, or having breakfast with our family, classifying the situation is easy

and if language choice varies with the situation it is clearly the situation that

decides the language not the other way round.

2.5.6 CS as Identity Relationship with Language

McClure (1977) deals with CS in the bilingual Mexican American

children. He finds that the language alternation marks a shift in identity-

relationships. It is more common among the older children than the younger

ones mainly because “the former have access to more identities” (McClure 1977:

105). Their knowledge of English increases with age, thus bringing about an

increase in the number of identity relationships associated with English.

Regarding topic, McClure feels that it does not have as much an influence in the

choice of the code (Spanish/English), as do the participants: “The children are

able to and in fact do, converse about any thing in their experience in both

languages. When a topic which is habitually discussed in one language happens

to come up in a conversation in the other language, there is high incidence of

code-mixing and code-changing.”

71

2.6 Studies of CM/CS in Indian English

Most of the studies on CM in the Indian Context, concentrate on code-

mixing with English in Indian Languages which is a Pan Indian phenomenon

(Ashok kumar 1995, Annamali 1989, Kachru 1983, Verma 1976, Jyostna 1980,

Chengappa 2005, Agnihotri 1998). There are a few studies on code-mixing of

Indian languages into Indian English. (Shastri 1988, Kachru 1983, Malik 1994).

2.6.1 Kachru’s Studies of CM/CS

2.6.1.1 Speech Function

Kachru (1966) has studied the socially determined speech functions such

as modes of address or reference, greetings, blessings, prayers, abuses, curses as

being related to the Indian context of culture. In Indian languages, there are fixed

format exponents for these contexts; but these are sometimes transformed to the

L2 for those contextual units that are absent in the culture of the English

language. Further, Kachru (1978 a) argues that a linguistic divergence is a result

of CM. The divergence may take a more extreme form, as in Hindi and Urdu or

Dakkhini.

2.6.1.2 CM as Lexical Innovation

Kachru (1975: 62-63) makes a distinction between „lexical Innovation‟

which includes the use of single lexical items and hybridization items (in South

Asian English) and assimilated items or non–restricted items in the native

varieties of English. For him, however, hybridized item is simply a lexical item

which comprises of two or more elements at least one of which is from a South

East Asian language and one from English. This includes both open set and

closed system items of lexical and grammatical elements.

Kachru‟s study on code-shifting is restricted to lexis. His is a data–

oriented study and as such, both the data and the treatment are limited to some

twenty-two imaginative writings by ten writers and some press materials chosen

from twelve Indian Newspapers; the treatment is in terms mainly of

„hybridization‟ at the lexical level. He comes to the conclusion that „CM and CS

need to be studied in a functional-context, both cross-linguistically and cross-

culturally, and that „the Indian subcontinent provides a substantial data for the

72

study of code-mixing both diachronically and synchronically.‟‟ (Kachru 1983:

206)

2.6.1.3 Context of Situation

Kachru (1978) presents a theoretical framework which would relate the

formal and functional aspects of such „language-mixing‟ and view it in a

pragmatic perspective in terms of the linguistic needs of the speech community.

He follows mostly Firth‟s framework. In Firth (1957a and 1957 b) and later in

somewhat modified form in Mitchell (1957), Halliday (1959), Ellis (1966) and

Kachru (1966), a schema has been presented towards delimiting texts with

reference to their contextually relevant categories. Firth has suggested the

following categories for the context of situation of a text:

A. The relevant features of participants: persons, personalities.

i) Verbal actions of participants

ii) Non-verbal actions of participants.

B. The relevant objects

C. The effect of the verbal action.

The other features to be considered are –

i) Economic, religious and social structures to which participants

belong.

ii) Types of discourse : monologue, narrative

iii) Personal interchange: age of participants.

iv) Type of speech- social flattery, cursing, etc.

CM is a role–dependent and function–dependent linguistic phenomenon.

In terms of role, one has to ask who is using the language, and in terms of

function, one has to ask what is to be accomplished by the speech act. In terms

of role, then, the religious social, economic and religious characteristics of the

participant in a speech act are crucial. On the other hand, in terms of function,

the specialized uses to which the given language is being put determine the CM.

In a sense, then, in several linguistically relevant situations there is a mutual

expectancy between the formal characteristics of the language (in this case, a

code-mixed language and its function). Kachru (1978) discusses the

73

phenomenon of CM in the theoretical framework of the „context of situation‟.

The concept, context of situation, provides a framework for relating language

use and linguistic form to the „immediate‟ linguistically relevant situation and

also to the „wider‟ context of culture.

A „contextual‟ unit means those features of a text which contribute to its

being assigned to a particular function. These features may be termed the

contextual parameters of a text. These would comprise linguistically relevant

clues, such as the participants, their sex, their positions on the social, caste or

religious hierarchy.

In certain contextual units, a multilingual person has the possibility of

choice between code-mixed (say Hindi and English or Persian) or non-code,

mixed languages. In such situations, the selection of a particular „code‟ is

determined by the attitude of a person toward a language (or toward a certain

type of code-mixing) or the prestige which a language (or a type of code-mixing)

has in a speech community.

Formal appropriateness in CM may be judged by using the concept of

formal cohesion. In other words, a particular type of lexical and grammatical

cohesion is associated with a specific type of discourse or register. In linguistic

terms, CM involves functioning at least in diasystem and as a consequence,

developing another linguistic code comprising formal features of two or more

codes. A linguistic code developed in this manner then develops a formal

cohesion and functional expectancy.

One might then say that the function of the CM languages is between

what is termed „digloassia‟ and CS. In the „diglosssia‟ situation, there is a

situationally determined use of two codes, and the codes involved are

functionally mutually exclusive. In CS, on the other hand, the functional

domains of the languages involved are determined by linguistically pluratic

situations, for example- the Punjabi-Hindi Code-switching in Haryana.

2.6.1.4 Mixing of English and Indian Languages

Kachru (1983) has argued that CM once considered linguistic sin has now

come to be recognized as a historically and sociologically determined

74

phenomenon with well-defined linguistic, pragmatic and attitudinal functions.

Several mixed languages have developed as result of this process. He discusses

CM with English in Indian languages and proposes ways of structuring the

phenomenon (Kachru 1975, 1978, 1983).

As much of his work deals with lexical items, he uses CM as a cover term

for CM/CS expressions. According to him, there are four broad types of mixing

seen in Indian Languages: 1) Englishization 2) Sanskritization 3) Persianziation

4) Pidiginization.

Regarding the pragmatic functions of the code-mixed varieties in Indian

languages, he notes that CM in Indian languages shows register identification

(register-wise mixing), e.g. Englishization (mixing English words),

Persianization (Persian words mixed in legal register– in lower courts), Literary

Criticism and philosophical writing (Hindi-Sanskrit words mixed). It provides

formal clues for style identification (Sanskrit, Persian and English). It is used as

device for elucidation and interpretation. It is possible when terminologies of

that language have not been stabilized or have not received acceptance. A person

uses two linguistic sources in defying concepts or a term so as to avoid

vagueness or ambiguity. It is used to redefine in other language (English) what

has been already expressed in one language (Hindi). Sometimes mixed lexical

items do not provide contextual clues and thus, language is used to conceal

various types of identities. In order not to give away caste identity, a person

prefers the English Kinship term brother–in-law, for the Tamil maccan or

attimbeer. CM is a contextually determined device; therefore there is mutual

expectancy between the type of CM and the contextual unit in which it

functions.

With this explanation he further claims that in register identification and

style identification, CM functions as foregrounding, because it is used to attract

attention. On the other hand in neutralization, it is used for the opposite effect,

i.e. to conceal the identity.

While observing the attitudes towards the functions of the code-mixed

varieties, he points out that, in the multilingual settings, all the code-mixed

75

varieties do not evoke identical attitudinal responses. A multilingual person

seems to choose CM of various types, pragmatically and attitudinally. He

considers the attitudes towards four types of mixing mentioned above: 1) Code-

mixing with English is a Pan-South Asian phenomenon. In attitudinal functional

terms, it ranks the highest and cuts across language boundaries, religious

boundaries and caste barriers. It is a marker of 1) modernization 2) socio-

economic position 3) membership in the elite group. It makes deliberate style.

The widest register range is associated with CM in English. It continues to be

used in the contexts such as- to demonstrate authority, power and identity with

the establishment. The evidence of this attitude is seen in the parents‟ language

preference for their children. It is also seen the in the choice of preferred

language in the college. 2) The attitude towards sanskritization is seen in

oratorical style associated with the rightist–Arya Samaj, whereas attitude

towards desanskritaziation is seen in Dravid Communities‟ oratorical style 3)

Persianization spread in those parts of India that came under the Muslim domain,

is associated with the legal register of the lower courts. The attitude towards

Persianization varies from one part of India to another. CM with Persian is used

in educational system in Kashmir or north India by Hindu Kashmir is and by

Muslims. But CM with Persian in South India is different. In Karnataka, Sridhar

(1978) observes that the educated Kannada speaker mixes English in his

Kannada, and the illiterate (earthly) one mixes Perso-Arabic in his Kannada. 4)

Pidginization which is low on the hierarchy is an attempt toward simplification

of language used in situation. Here, the participants speak languages which are

not mutually intelligible. This mixed type variety is called Bazar Hindi, Butler

English or chi chi English (Kachru 1987 a).

The formal characteristic of CM texts is their lexis and lexical cohesion

(Halliday and Hassan 1976 in Kachru 1983) “A user of a code-mixed variety

intuitively applies the process of the first language to nativise the linguistic

elements of the other code. In Hindi-English code-mixing, most of the

productive grammatical processes of Hindi-Urdu are applied to English items.

76

For example, company-companiya, master-masterin, doctor-doctarin.” This also

applies to inflection assignment and other grammatical categories.

There is a type of cline in mixing which starts with lexical mixing and

then progressively extends to higher units, the maximum being an alternate use

of sentences from two codes. The mixing at lexical level may show a lexical

spread which is associated with a register (e.g. paragraph with Persian Lexical

items typical of legal register in Indian language) (Persian +Hindi); on the other

hand, there is almost an alternate use of units of Hindi and English (e.g.

paragraph with switching between Hindi and English).

2.6.1.5 Code Mixing vs. Odd Mixing

About unacceptable CM, Kachru (1978b) comments that CM is both

functionally and formally a rule governed phenomenon. It is not an open-ended

process, but has various collocational and grammatical constraints. There is a

point both in grammar and lexis when a user distinguishes between code-mixing

and odd-mixing, The responses to the code-mixed items by the users of such

varieties seem to vary from „yes, acceptable‟ to „no unacceptable‟ ,‟well

depends‟, and „I don‟t know‟.

2.6.1.6 Impact of CM on Hindi Syntax

There are several studies that concentrate on the phonological, lexical and

syntactic aspects of Hindi with mixing of the Persian words, Hindi with mixing

of the English words. Kachru (1983) summarizes the researches dealing with the

impact of CM and CS on Hindi syntax:

1) There is change in word order. The preferred word order of Hindi is

SOV as opposed to the SVO of English. In creative writing in Hindi and

newspaper register, we find example of SVO too. This may be due to the

influence of an English substratum (Mishra 1963 in Kachru 1983) and the

practice of fast translation from primarily English texts in journalism

broadcasting.

2) There is the introduction of indirect speech. In Hindi discourse,

traditionally, no distinction is made between direct and indirect speeches.

77

But modern Hindi prose construction such as „NP said that he will read’

as against „NP said that will read’

3) There is the use of impersonal constructions. For example,‘ it is said’ ,

‘it has been learnt’, ‘It is claimed’, type of constructions. In Hindi

newspaper, it is „kahajata hai’, - is now being used.

4) In the Indo–Aryan languages, in general, there is a tendency to delete the

agent in passive communications. In new Hindi, Dwara, Jariye, are used

because of the influence of English.

5) The use of post-head relative clause with „jo’ is also attributed to the

influence of English and by some to the influence of Persian. For

example, “Voh larka jo table par baitha hai mera bhai hai.” (That boy who

is sitting on the table is my brother).

6) There are two views regarding the use of the parenthetical clauses. For

some, they are the result of the English influence and for others; they are

typically Indo–Aryan constructions.

2.6.1.7 Exponents of ‘Mixing’

The formal exponents of „mixing‟, as stated in Kachru (1978 a), form a

hierarchy. In this hierarchy, mixing of simple lexical items rank the lowest and

the mixing of sentences rank the highest. They are NP insertion, VP insertion,

unit hybridization, sentence insertion, idioms and collocation insertion,

Inflection attachment and reduplication. This process includes the use of

reduplication of English items. In some south Asian languages, reduplication has

the function of marking identification, e.g. petrol vetrol , acting vekting, Taim

vaim , kar var.

2.6.2 Shastri’s Study of CM

Shastri‟s contribution to Indian English is worth noting, especially regard

to the compilation of the two corpora on Indian English: Kolhapur corpus of

Indian English (1988) and ICE-IND (2002). His research paper (1988) „Code

Mixing in the Process of Indianization of English‟ is based on Kolhapur Corpus

of Indian English. As our present study is based on CM and CS in Indian English

78

speech and as it is based on ICE-IND, we have considered his paper here in

some detail.

2.6.2.1 Code Mixing Process

Shastri (1988) considers CM of Indian languages into IE as an important

contributory factor in the process of Indianization of English. He considers

„hybridization‟, „absorption‟ and „assimilation‟ as distinct stages, in that order, in

the process of borrowing in languages. According to him, „absorption‟ is a

process of naturalization (the borrowed items which become stabilized by

gaining linguistic and socio-culture sanctions at the local or regional level). The

language speaking community tacitly accepts it. The term „assimilation‟ is used

for those that become part and parcel of the native varieties of English. The need

for assimilating Indian elements into English has once again arisen especially in

the Western World and in the fields of Fine Arts to cater for the desire for

International understanding though cultural sympathy. So, a number of register-

bound items have already found their way into standard British and American

dictionaries. More striking forms of assimilation are those that borrow the

abstract concepts of Indian socio-cultural phenomena and the Indian way of

conceptualizing reality through the Indian language items.

2.6.2.2 Functions of CM in Indian Context

Kachru (1983) has suggested the pragmatic and the attitudinal functions

of CM. The first manifests in the register-determined contexts and the second as

markers of the social status and attitudes. Shastri‟s (1988) study has revealed a

third aspect, i.e. a deliberate attempt to Indianize English, indicative of the

national aspiration typical of all the newly independent nations which expresses

itself in different ways in different nations. This deliberate attempt is more like

the early American efforts at the Americanization of English.

Moreover, it is a deliberate attempt not only at asserting the newly free

nation‟s distinct identity as strongly and emotionally as possible, but also at

adapting the English language so as to bring it within the reach of an average

educated Indian, a truly „socialistic motive‟. This may well be labeled as a

feature of pan-Indian English. Such attempts may range from genuinely

79

nationalistic to ridiculously parochial innovations. He gives a few examples:

Poorna-Swaraj, Azadi, Swadesh, Kala Parishad, Askashwani, Doordarshan,

Sangeet Natak Akademi, Shri Shrimati, and Sarvashri. There are quite a few

new items, the most talked about being- bundh, hartal, gherao, not to speak of

Dravida Munnetra Kazagam, Shiv Sena, Hindu Ekata etc. - the kind of which

have found their way into pan-Indian English, a kind of lingua franca typical

not only of the press but popular English literature in India and the bulk of

spoken English for everyday communication.

With regard to the formal treatment of CM, Shastri notes that, CM does

not confine itself to any particular level of language structure. It tends to affect

the code-mixed language at all levels. In phonology and morphology, it is most

obvious or transparent; but at the level of syntax and specially semantics, it

tends to be not so obvious or opaque. His corpus-based study deals with the

structural variety of CM in detail and also notes some examples of exponents

and constraints.

2.6.2.3 Structural Variety and Stylistic Motivation of CM

A) The details of the structural variety of CM marked in his study are as

follows:

1. Basic NPs

a. N +N Constructions in which the modifier N is an Indian item:

Tehsil level =level of the tehsil (administrative area) (All)

b. N+ N Constructions in which the head N is an Indian item:

Block Samiti= Committee to manage the block (All)

Here, the head nouns may be inflected the English way: as for example,

deeyas.

c. Adj=N Constructions in which the adjective is an Indian item:

Peria colony = big colony, really colony of the big (A09)

Here, too, the adjectives have their derivational affixes, swadesh, jirki

d. Quantifier + N constructions in which the head N is an Indian items. :

Ten thousand paras (measure) (KO1)

Notice the last phrase which is very English phrase.

2. Basic Coordinated NPs

N and N constructions in which either of the Ns is an Indian item:

AIR and Doordarshan (TV)(CO9)

80

3. Complex NPs with multiple premodification:

a. All modifier Ns :

Siasm sag timber (All)

b. Ns and Adjs

Purna Swaraj pledge (A10)

4. Complex NPs of sorts :

Badawar Tai (poor people’s mother) Indira Gandhi (A28)

5. Attributive constructions

a. N + Adj. constructions in which the N is an Indian item:

( mehandi-red (K10)

b. N + V-ed constructions in which the N is an Indian item:

Rishabha-based (CO2)

c. N+ V-ing constructions also as substantives in which the N element is an

Indian item.

Chappal- throwing (B22)

6. Inflected Forms : Indian items with English inflections :

A. Plurals in –s (simple forms)

By far the most productive English inflection used is the –s plural inflection:

Hamals (A09), harijans (A22)

B. Plurals in –s (complex forms)

Kabir panthis (A42)

C. Adjectives in –te and –ist

Baindavanite (CO5), harmonist (CO4)

D. Abstract nouns in –ism :

Brahmanism (B10), Hinduism (C10)

E. The –s Genitives :

a. Possessive Genitive : Prana‟s connection with Prajapati = Prana has

connection with prajapati (D05);

b. Group Genetive: raja of Dilwar’s cousin = the cousin of the raja of

Dilwar (KO5).

7. Prefixes and Suffixes : trans-Jamuna (A37), filmwalahs (L05)

Some of these features have no doubt been pointed out earlier by Kachru.

But the analysis presented here not only confirms some of his hypotheses, but

also indicates how CM in IE is a far more complex phenomenon affecting the

very fabric of the language at all levels of delicacy, for example- the structural

variety of the –s genitive in particular.

He uses the term stylistics in the broadest sense of the term to include what

the speaker wants to accomplish in a given context of situation.

81

B) Stylistically Motivated Code-Mixing:

Exponents of Code Mixing:

1. Single lexical items with restricted/extended meanings :

i) …mass sterilization camps… which are little more than melas. (B14)

ii) she would keep some of ….some of her charm for those …. who

might go on the line after the tamasha (minister’s visit to the slum) was

over. (K18)

No. i) is an illustration of exploitation of the associated meaning (-DESIRABLE)

of the item mela which literally means a „fair‟. No. ii) is an example of a

similar phenomenon, the item tamasha which could have an associated

component (+CONTEMPT) which otherwise literally means different things in

different Indian languages. The example seem to indicate that there is a need for

recognizing a fuller range of stylistic marking potential of the code-mixed items

even at the single lexical item level.

2. NP insertion

There are instances of Indian language NPs inserted into English with or

without English paraphrases. The writer‟s decision to supply or not to supply

paraphrase may be dictated by the audience readership he has in view or the

desire to compensate for the inadequacy of one of the phrases.

For example: The succeeding thumr ki bandish based on „khamaj’,

‘Adana’ and ‘Bahar’ and the final „dadra’ were all

redolent of the distinctive purab flavor (C04).

It Fills a gap in the receiving language, but could have been phrased as

the bandish of thumri; the motivation seems to be one of „Cohesion‟ or may be

of constraint.

3. Sentence Insertion

a. Without paraphrase:

Boja babuji? (Any load to carry Sir?) Those two words sent a shiver

through my body. „ Chalo, hamare sath’ ( Come with me) I

commanded her (K52)

b. With paraphrases :

Dramatic performance becomes, according to „Naatya Adriana’ an

art when it arouses sentiments in the minds of the audience (Sabha

yaanaam hridayam nartayati iti naatakam) (= That which makes the

audience’s heart dance is a drama)(C07).

4. Modes of address

i) „Yes, yes baba I am moving. I am moving’ and she tottered up the

aisle of the bus. (K50)

ii). „who is speaking? Jaydeep here. Yes Jaydeep, I am bit in a soup

dafa. Won‟t be able to come –yes.‟(L01)

These examples of address- baba (daddy, meaning any one who assumes

the role of one), Mai, ma (mother, one as good as mother), Beta (Son, one as

good as son), Deva (God-Hindu), Inshaalla (God-Muslim), Atte (aunty –regional

–Karnataka) suggest the wide range of modes of address with unmistakable

social function.

82

5. Interjections:

i) I bit a little pedha so that the remainder appeared, in my palm like

the moon a few days before becoming full. „Cha’ I cried, spitting

out what I had eaten. (K34)

ii) Choma heard the landlord‟s mother curse form inside the house

„Abba, the insolence of these holeyas‟. (K03)

The interjections Cha! Chee! Chee! are pan-Indian expressions of strong

disapproval: while Abba is limited to south Indian expressing shock at „temerity‟

– the limit.

6. Idioms and Proverbs:

This is all too common a phenomenon – the use of idioms, proverbs and

sayings which are culture-bound and, as rule, untranslatable. Occasionally,

parallel idioms occur in different languages. But the fact remains that these are

the surest markers of socio-cultural traits.

Regarding „constraints‟ on CM, this study suggests that there are two

kinds of constraints on CM: First, as discussed by Kachru (1978), the relevant to

the cline of acceptability, and the second, the language specific constraints. This

phenomenon seems to operate at the semantic level and, therefore, is likely to be

reflected in the processes of lexicalization and affixation or the realization of

open set items and closed-system terms.

Shastri refers also to the near absence of mixing of verbs in Indian

English. For example, Gherao (peaceful intimidation by crowding in on

person(s)), in Hindi is a noun, one has to add an operator karna to derive a verb

i.e. gherao karna . In the form gheraoed was obviously treating the item the

gherao is treated as verb with the English past –ed inflection. He observes that

there is a general tendency in Indian words to follow the English grammatical

processes of affixation for realizing grammatical meanings. Only in the case of

film-wallah, the affix wallah is from Hindi; but the root film itself is from

English and the item filmwallah is borrowed back into IE. Barring such

examples, he concludes surmises that the mixed items normally conform to the

morphological patterns of the receiving language.

83

2.6.3 Verma’s Study of Socio-cultural Factor in CS

Verma (1976), in his paper „Code-Switching: Hindi-English‟, begins with

a distinction between linguistic competence and communicative competence.

For him, “Linguistic competence does not necessarily mean communicative

competence; but communicative competence subsumes linguistic competence,

for it may be described as linguistic competence plus situational appropriateness.

Linguistic competence is abstract, context-free, rule governed linguistic

behavior; communicational competence is context-governed topic-oriented

externalization of linguistic competence.”

This paper attempts to illustrate certain aspects of Hindi-English CS

among the educated speakers of Hindi-English. English and Hindi have

coexisted within the same linguistic area as part of same socio-economic

complex for years. Both the codes have been serving important function in the

daily social interaction. These speakers use different varieties of their mother

tongue in various situations in life; but when they have to use the technical

register, they normally switch over to English. This kind of register-oriented

bilingualism is labeled as „registral bilingualism‟ (Verma 1969:302). In fact, the

situation is never as simple as this.

The chief regulations of Hindi-English CS or of mixture of Hindi-English

styles are related to the level of education and the topic of discourse. The higher

the level of education and more technical the topic of discourse the greater the

degree of mixture and frequency of switching. It may, therefore, be said that the

bilingual switching is patterned and predictable on the basis of topical and

situational feature. He explains the dominance of Hindi/English by way of cline.

At one end and in certain roles, these bilinguals use only English and at the other

end and in certain other roles, they use only Hindi. English is used in highly

formal situations to talk about technical topics; Hindi is used for intimate,

informal, personalized statements. In between these two extreme ends, we find

different degrees of English-dominated and Hindi-dominated mixtures.

84

The cline may be shown as:

English English dominated Hindi dominated Hindi

CS in multilingual settings is regulated by the topic of discourse and

stylistic (formal/informal) considerations. In this regard Fishman (1972: 439)

observes, “The fact that two individuals who usually speak to each other

primarily in X nevertheless switch to Y when discussing certain topics leads us

to consider topic per se as a regulator of language used in multilingual settings.

The implication of topical regulation of language choice is that certain topics are

somehow handled „better‟ or more appropriately in one language than in other in

particular multilingual contexts”. The speakers do not radically switch from one

style to another; but they build on the co-existence of alternative forms to create

meaning. Switching from one code to another is not a matter of free individual

choice. It is affected by topical and situational features, which determine the

speaker‟s choice from among a set of available codes.

Regarding the „context‟ in CS, Verma observes that the CS is context-

governed. If the interlocutors have a degree in English and use English as their

medium of lecturing or in their office work, they are likely to use information

carrying items of English and linkers of Hindi. If their topic of discourse is

technical, their registral items are likely to be from English and the grammatical

items from Hindi. CS, one might say, is a marked badge of educated, urban

bilinguals. The preponderance of a particular set of lexical items of a code, he

notes, also depends upon our emotional and intellectual attachments to the code.

2.6.4 Ashok Kumar’s Study of CS and Motivation

Ashok Kumar (1987) focuses on „motivation‟ for CS. He considers, apart

from the communicative necessity on the part of the bilingual, there are some

more reasons which are equally if not more effective in explaining this

phenomenon. These have either been superficially dealt with, or have been

totally ignored. His study of the Hindi-based bilinguals, both educated and

uneducated indicates that the alteration from Hindi to English is dictated also by

such considerations as a)Switching under emotional stress, b) Switching for

85

imposing authority, c) Switching for fashion, d) Inevitability, e) Technical CS, f)

Switching for business, g) Switching in creative writing and h)Euphemistic CS.

He deals with some of the important formal and functional features of

Hindi-English code-switching. The analysis of the linguistic aspects of Code-

switching refutes some of the code-switching constraints. The functions of code-

switching are more or less in line with Gumperz (1982), although they have been

dealt with more elaborately in his study.

2.6.5 Other Studies of CM/CS in Indian English

Malik Lalita (1994) discusses the role of Code-switching in the formation

of Indian English. On the basis of her investigation, she asserts that Code-

switching in bilingual or multilingual speech is not a „grammarless language

mixture‟. On the contrary, morphological, semantic and pragmatic analyses

show that there are marked linguistic, semantic and attitudinal functions of Code

Switching. The process of CS plays a major role in the formation of Indian

English; its contribution to the formation of IE is, no doubt, noteworthy. This

study traces also the educational implications of Code-switching in Indian

English.

Jyostna (1980) studies the formal functions of CM in Indian films. Her

study is based on the data collected from the spoken media. She argues that CM

may occur at a variety of grammatical levels. CM obeys certain syntactic

constraints. It is restricted by social aspects; however, as she insists, CM has

certain pragmatic advantages.

The work of Nirmale (2009) „A study of CS and CM in the selected

fictional works by Indian Writers in English‟, critically examines and investigate

the phenomenon of code mixing and code switching as manifested in Indian

writing in English in general and fictional work of M.K. Anand and Raja Rao,

Farukh Dhondi and Shobha De‟ in particular. The study probes into the process

of nativization which mainly operates through the mechanism of CM and CS.

To conclude the discussion on Code Mixing and Code Switching, it may

be observed that CM and CS have over years been the areas of great concern for

the linguists and the researchers. Their scope is not limited to one language or

86

medium, one region or community, one field of knowledge or communication.

In fact their scope is extended over to several languages and mass media used for

information and communication by users all over the world. A number of

linguistic, socio-cultural and pragmatic factors contribute to their formation and

usage. That also explains the researchers‟ keen interest in the study of CM and

CS in Indian English.